Answer: Purchase Compute Savings Plans for the predicted medium load of the EKS cluster. Scale the cluster with OnDemand Capacity Reservations based on event dates for peaks. Purchase 1-year No Upfront Reserved Instances for the database to meet the predicted base load. Temporarily scale out database read replicas during peaks.

B is the most cost-effective setup for the platform due to the combination of several key cost-optimization strategies: - **Compute Savings Plans for EKS**: Compute Savings Plans allow you to commit to a consistent amount of compute usage ($/hour) over a 1 or 3 year period and benefit from lower prices across Amazon EC2 and AWS Fargate. This makes it standard to choose this for the predicted medium load, leveraging consistent usage patterns to achieve cost savings. - **On-Demand Capacity Reservations**: These reservations are more flexible compared to other reservations. They ensure that you have the capacity you need, but only pay for them when you use them. Given that peaks depend on event dates, this option allows you to scale up only when needed without long-term commitments, reducing costs during off-peak times. - **No Upfront Reserved Instances for RDS**: Choosing 1-year No Upfront Reserved Instances for the database to meet the predicted base load strikes a balance between cost savings and flexibility. You don’t have to pay anything upfront and you get a discounted hourly rate, making it financially easier to manage. This is ideal for a component like a database where base load expectations are stable. - **Temporary Read Replicas**: Scaling out the database read replicas during peaks ensures the database can handle increased read traffic, which is typical for activities like event ticketing during high demand. Read replicas can be added and removed as needed, leading to cost savings when demand is low. This combination allows for consistent savings on predictable workloads while efficiently managing and minimizing costs during demand surges.

Author: LeetQuiz Editorial Team